Customer Data Platform: 4 Reasons to Invest in a Readymade Solution Instead of Using Your Data Warehouse as Your CDP

A Customer Data Platform comes in many flavors making it hard to know when it’s a good time to invest in one. This guide tells you when you need a CDP and when you don't.

By Arpit Choudhury on

Data

Introduction

With Segment’s recent acquisition by Twilio for $3.2 Billion, one would expect more consolidation in the market for Customer Data Platforms (CDPs). While we might see more mergers and acquisitions going forward, the industry is showing no signs of slowing down and the number of companies offering Customer Data Platforms is only on the rise.

This is a clear indication of the fact that Customer Data Platforms are experiencing rapid adoption and are here to stay. But does it always make sense to invest in a readymade CDP instead of building your own?

Certainly not! Here’s a strong argument for using your Data Warehouse as your Customer Data Platform.

However, in many cases, it does make sense to opt for an off-the-shelf solution and we want to highlight the top 4 reasons to choose one.

  • You operate in an industry that can benefit from purpose-built or vertical solutions that cater to your needs
  • You want to empower your marketers to consolidate customer data from different sources and act upon them via the no-code capabilities of a CDP
  • You don’t intend to or lack the resources to invest in a minimum viable data stack (MVDS) comprising a data warehouse (Snowflake, Google BigQuery, or AWS Redshift), an ELT-based data integration tool (Fivetran, Stitch, or Matillion), and a Reverse ETL tool like Hightouch
  • You want to leverage real-time capabilities offered by some readymade CDPs

Let’s take a closer look at each of these.

You need an industry-specific Customer Data Platform

Industry-specific or Vertical Customer Data Platforms are those that are built to serve a specific purpose or the needs of a particular industry.

Amperity and Zaius are examples of such vertical Customer Data Platforms.

  • Amperity is focused on the needs of consumer brands
  • Zaius solves the needs of ecommerce stores

If you operate in one of the above industries and are looking to tackle problems that can be elegantly solved by a readymade Customer Data Platform, it makes complete sense to adopt one.

That said, it is still imperative to use a data warehouse to store all the data that moves through the CDP. Doing so will ensure that you own your data so that if and when the need arises, you can switch to another vendor or a warehouse-centric solution.

You want to empower your marketing team

It is fair to say that Customer Data Platforms have out-of-the-box capabilities that give marketers the ability and flexibility to solve a lot of their problems without relying on engineers or data people.

Data unification and Identity resolution — a core offering of Customer Data Platforms — is a hard problem to solve unless you have dedicated data engineers who have the requisite skills to handle this via SQL.

Since readymade CDPs consolidate customer actions across different touchpoints and build unified customer profiles, marketing teams are able to build segments by combining customer attributes and actions and export these dynamic segments to third-party tools used for growth and advertising — all without the need to know how to write code or SQL queries.

Additionally, marketing and growth teams get to have more control over their data needs and can easily configure new sources and destinations that a CDP offers pre-built connectors with. What usually takes weeks for data or engineering teams to implement via code can be done in a matter of hours by semi-technical folks.

However, there are limitations to what you can do on top of the data model of a CDP which cannot be bent to mimic custom data models — something that is needed more often than not.

For B2B SaaS products, it is common (and a best practice) to track and act upon data about users as well as accounts (also referred to as organizations or workspaces). An account comprises multiple users but a user can also own or be part of multiple accounts, often using the same email address.

A readymade Customer Data Platform centers around the customer or user and therefore cannot elegantly handle the aforementioned many to many relationships between users and accounts. This is something worth keeping in mind, especially if you offer a B2B SaaS product.

You’re not ready to invest in a Minimum Viable Data Stack

An MVDS comprises the following:

  • An ELT-based data integration tool such as Fivetran
  • A data warehouse such as Snowflake
  • Lastly, a Reverse ETL tool like Hightouch

Needless to say, implementing the above is a significant investment both in terms of money and manpower. Therefore, if you don’t intend to do this anytime soon and your marketing team is still looking for ways to combine and move customer data to third-party tools, it makes complete sense to use a readymade Customer Data Platform.

That said, if you’re not warehousing your data already, you need to do that sooner rather than later so that your data is not locked away with your CDP vendor. All leading Customer Data Platforms integrate with data warehouse providers like Snowflake, Google BigQuery, and AWS Redshift, and getting one up and running has never been faster and cheaper.

After all, at some point, you’d want to adopt a warehouse-based approach to data integration which is only as good as the data in the warehouse itself.

P.S. Check out our guide on the data integration landscape if you’re curious to understand the differences between ELT, CDP, and Reverse ETL.

You need the real-timeyness of a Customer Data Platform

Most Customer Data Platforms offer real-time data processing capabilities and have become pretty good at processing large volumes of data very quickly. Therefore, if you have use cases that rely on real-time capabilities, a CDP is the way to go.

That said, it makes sense to take stock of your real-time data processing needs before you decide to invest in a Customer Data Platform. To do this, try to answer the following question:

Do you need to personalize the core user experience in real-time based on how a user interacts with your product?

A relatable example is how Uber or Lyft calculate the fare for a ride by combining several data points in real-time. Although at their scale, building a Customer Data Platform will make more sense.

Or perhaps you want to deliver a welcome email to users as soon as they sign up for your product and offer a more personalized experience by customizing the email based on the user’s industry and role.

In such cases, the real-time capabilities of a Customer Data Platform are very handy. That said, keep in mind that a bulk of general marketing use cases such as sending emails or showing ads based on user events don’t need real-time processing of data as these follow-up actions don’t really happen in real-time.

It’s worth mentioning that often companies evaluating a Customer Data Platform might not have an immediate need for the core capabilities of a CDP, and this is especially true for companies that are looking to set up a data stack from scratch. So here’s why you might not need a CDP at all

If your goal is to just track customer data and send it to sales, marketing, and analytics tools, you don’t need a Customer Data Platform at all. Segment is the most popular choice to do this but essentially, all you need is Connections which is one of the products Segment offers. Personas is their CDP offering which is available as an add-on.

Needless to say, Segment is not the only choice here — Snowplow Analytics and Rudderstack are also popular alternatives that are good at event tracking.

Additionally, instead of piping customer data directly to third-party destinations, there are many advantages of storing the data in a data warehouse first. Data can be transformed (to meet the schema requirements of each destination) and enriched with data from other sources.

This is also where a Reverse ETL tool like Hightouch comes in handy as it can be used to sync the clean, enriched data from the data warehouse to sales, marketing, and analytics tools. This approach also results in your data warehouse becoming the source of truth for your customer data.

Conclusion

There are several other reasons to choose a readymade Customer Data Platform but the four mentioned here are the core ones. It’s likely that at least one of these resonates with you but if by chance that’s not the case, you need to reconsider investing in a CDP because implementing one and doing it well is a significant undertaking that can take weeks if not months.

If you’re still on the fence and are unsure if a Customer Data Platform is right for you, schedule a chat with us and we will be happy to help you evaluate the right solution. And rest assured, we will not pitch our product unless we see a strong use case for it.